Abstract
BACKGROUND: This meta-analysis aims to provide an update on the efficacy of individual cognitive behavioral therapy (CBT) for eating disorders (EDs) in the light of recent advances in the field, particularly the emergence of new approaches like self-help. METHOD: We conducted multivariate multilevel meta-analyses using data from 42 randomized controlled trials to evaluate the efficacy of individual CBT for anorexia nervosa, bulimia nervosa, binge-eating disorder, and mixed eating disorder samples. Outcomes included ED pathology, the frequency of binge eating and compensatory behaviors, and body mass index. The type of comparison group, treatment format, treatment duration, baseline ED severity, and year of publication were analyzed as potential moderators. RESULTS: CBT was more efficacious with regard to all relevant outcomes than waitlist conditions. For anorexia nervosa, there were no studies using a waitlist condition. CBT was not superior to active treatments for any disorder or any outcome. Therapist-led treatment for bulimia nervosa showed a significant effect for all outcomes, while self-help formats only did for some outcomes. Treatment duration and publication year were significant moderators for mixed EDs, while greater baseline severity was associated with larger effects of CBT for binge-eating disorder. DISCUSSION: The variety of active controls impeded comparative analyses of CBT with specific active controls. Effects were heterogeneous; the majority of studies had a poor risk of bias assessment. There is a paucity of research, particularly of treatment studies for anorexia nervosa.